A New Data Set of Thermospheric Molecular Oxygen From the Global‐scale Observations of the Limb and Disk (GOLD) Mission

Abstract

The Global-scale Observations of the Limb and Disk (GOLD) instrument was launched on 25 January 2018 onboard the SES-14 commercial communications satellite and began nominal science operations in October 2018. Operating from geostationary orbit at 47.5°W longitude, GOLD images the Earth's thermosphere and ionosphere in the far-ultraviolet (132-162 nm), measuring critical geophysical parameters by continuously scanning the Earth's disk and limb 18 hours per day. GOLD also performs stellar occultation measurements using bright type O and B stars. Up to 10 occultations per day are obtained at latitudes ranging from 60°S to 45°N and two fixed longitudes,~33°E and~128°W. The occultation data provide a direct measurement of atmospheric absorption in the O 2 Schumann Runge continuum, which is used to retrieve the O 2 density profile from 130-to 200-km altitude. This paper describes the GOLD occultation measurement technique and the operational retrieval algorithm used to derive the Level 2 O2DEN data product. The spatial, temporal, and local time sampling of the GOLD O 2 data product is summarized, and the data quality, retrieval errors, and validation plans are discussed. Occultations are observed under both daytime and nighttime conditions, providing the O 2 density over a complete range of local times. GOLD retrievals have an estimated precision of 10% and a vertical resolution of 5-10 km, depending on altitude. The data have sufficient accuracy and resolution to provide scientifically useful constraints on the O 2 abundance and variability. Initial comparisons show qualitative agreement, but also notable differences, between the GOLD O 2 data and the Naval Research Laboratory Mass Spectrometer Incoherent Scatter Radar (NRLMSISE-00) empirical model predictions. Key Points:• Thermospheric O 2 density profiles derived from stellar occultation measurements by the GOLD instrument are described • The GOLD O 2 data set covers altitudes from 130 to 200 km, all local times and latitudes from 60°S to 45°N, with 10% precision and 5-to 10-km vertical resolution • Preliminary validation shows overall agreement with empirical models but also significant differences in magnitude and local time dependenceCorrespondence to:

Thermospheric Composition and Solar EUV Flux From the Global‐Scale Observations of the Limb and Disk (GOLD) Mission

“…During the remaining time, when the Sun is within 30° of the instrument's field of regard, GOLD is placed in Solar Safe mode. GOLD's routine observations consist of four different observation scenarios: dayside disk scans; limb scans (Evans et al., 2020); occultations (Lumpe et al., 2020); and nightside disk scans (Eastes et al., 2019). DAY disk scans begin at 03:00 satellite local time (06:10 UT) when the sun‐illuminated atmosphere first appears in the GOLD field of view and occur on a 30‐min cadence.…”

Neutral Exospheric Temperatures From the GOLD Mission

“…Second, we will increase the quantity of usable data products by improving the rate of successful convergences of the algorithm. This will be accomplished by replacing the use of Craig Markwardt's MPFIT routine with a more robust optimal estimation algorithm currently being used to retrieve O 2 density profiles from GOLD stellar occultation observations (Lumpe et al, 2020).…”

Initial Observations by the GOLD Mission

“…An example of the O 2 density profiles retrieved from nightside occultations on a single day is shown in Figure 8. Lumpe et al (2020) have described the occultations and retrieval of O 2 densities. The profiles retrieved for each of the five occultations (lines with diamonds) are compared with the retrieval a priori (smooth, black line) The GOLD O 2 algorithm uses a fixed a priori profile for all retrievals.…”

“…This profile is a derived (from the Naval Research Laboratory Mass-Spectrometer-Incoherent-Scatter model, NRLMSISE-00, Picone et al, 2002) model run for solar minimum conditions. As discussed in Lumpe et al (2020), the retrieval is tightly constrained toward the a priori at the highest (>200 km) and lowest (<130 km) altitudes, where the information content is low due to noise dominance (low O 2 densities) or complete absorption. Between 130 and 200 km, the O 2 density is of high quality with zero a priori bias.…”